Decay in the Eastern Tropical North Atlantic - ACS Publications

Aug 5, 2013 - Pathways of Superoxide (O2–) Decay in the Eastern Tropical North .... upper water column of the Eastern Tropical North Atlantic (ETNA)...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

Pathways of Superoxide (O2−) Decay in the Eastern Tropical North Atlantic Kathrin Wuttig,† Maija I. Heller,†,‡ and Peter L. Croot*,†,§ †

Marine Biogeochemistry, GEOMAR Helmholtz Centre for Ocean Research Kiel, Düsternbrooker Weg 20, 24105 Kiel, Germany Marine Environmental Biology, Department of Biological Sciences, University of Southern California, Los Angeles, California 90089, United States § Earth and Ocean Sciences, School of Natural Sciences, National University of Ireland, Galway (NUIG), Galway, Ireland ‡

S Supporting Information *

ABSTRACT: Superoxide (O2−: IUPAC name dioxide (•1−)) is an important transient reactive oxygen species (ROS) in the ocean formed as an intermediate in the redox transformation of oxygen (O2) into hydrogen peroxide (H2O2) and vice versa. This highly reactive and very short-lived radical anion can be produced both via photochemical and biological processes in the ocean. In this paper we examine the decomposition rate of O2− throughout the water column, using new data collected in the Eastern Tropical North Atlantic (ETNA) Ocean. For this approach we applied a semi factorial experimental design to identify and quantify the pathways of the major identified sinks in the ocean. In this work we occupied six stations, two on the West African continental shelf and four open ocean stations, including the CVOO time series site adjacent to Cape Verde. Our results indicate that, in the surface ocean impacted by Saharan aerosols and coastal sediment resuspension, the main decay pathways for superoxide are via reactions with Mn(II) and organic matter.

■

Previous work on O2− reactivity in filtered seawater has shown that reactions with organically complexed copper (Cu) are the dominant decay pathway.2,6 Hence reactions of iron (Fe) with O2− appear to be a smaller sink for O2−6,17 but may be significant for the formation of Fe(II) in seawater18,19 by reduction of inorganic or organically complexed Fe. More recently Mn was identified as a potential sink for O2− in surface waters.5 CDOM may also be a significant sink for O2− in natural waters.8,20 A schematic summarizing the different pathways for superoxide decay is shown in Figure S1 of the Supporting Information. The aim of the present work was to determine and quantify the pathways of O2− decay in the upper water column of the Eastern Tropical North Atlantic (ETNA) Ocean. We examine here the role of natural organic matter and the complexes of Cu, Fe, and manganese (Mn) in open ocean seawater to act as sinks for O2− by relating the decay rate of O2− to the metal concentrations and speciation. In the present work, the concentration of O2− in the experimental treatments was followed by a chemiluminescence technique using MCLA after additions of potassium superoxide (KO2) as described previously.6−8,17

INTRODUCTION The short-lived superoxide (O2−) radical is the intermediate reaction product in the formation of hydrogen peroxide (H2O2) from oxygen (O2) and is an important reactive oxygen species (ROS) in its own right in the redox cycling of metal ions and organic matter in the ocean.1−8 In the sunlit surface ocean O2− is predominantly produced via O2 reactions during the photooxidation of colored dissolved organic matter (CDOM).9 O2− may also be produced in the water column via metabolic processes in phytoplankton and bacteria.10−13 Other sources of O2− in seawater may be significant below the euphotic zone, and these may include the oxidation of reduced thiols and hydroquinones/semiquinones derived from lysis or breakage of bacteria and phytoplankton cells.7,8 There have been a few attempts to directly measure O2− production rates in open ocean seawater, with a single study published on photoproduction.9 More recent work has examined nonphotochemical production13−15 utilizing the reaction between O2− and 2methyl-6-(p-methoxyphenyl)-3-7-dihydroimidazol[1,2-α]pyrazin-3-one (MCLA) to produce chemiluminescence.16 For some metals, redox pairs exist (e.g., Cu(II)/Cu(I), Fe(II)/Fe(III), and Mn(II)/Mn(III)) where both species react rapidly with O2− leading to a catalytic cycle for superoxide decay.2,5,6 These metals form the redox center for enzymes such as catalases, peroxidases, and superoxide dismutases (SODs) that are present in organisms to protect the cell from damage from ROS. © 2013 American Chemical Society

Received: Revised: Accepted: Published: 10249

April 16, 2013 August 2, 2013 August 5, 2013 August 5, 2013 dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

Figure 1. Four different cases of the catalyzed disproportionation of O2− in seawater.

■

measurement was started; this is equivalent to an initial O2− concentration of 36 nM. The procedures were adapted according to our earlier work.6−8,17 Decay rates (Table S4, SI) were similarly assessed using our earlier procedures.6,17 In brief, the value of kobs was determined from the combined first and second order fit to the raw data using the Levenberg− Marquardt algorithm,22,23 and the error estimate (standard deviation) was evaluated from the χ2 result of the fit.24 The rate constant and the errors were then transformed into log k values using the classical propagation of random errors as described in Miller and Miller25 (σlogk = | σk (−0.434/k) | ). In this work we use the same terminology for the measured decay rates for O2− as used in our previous works: kSW (s−1), reaction in unamended seawater; kDTPA (s−1), reaction in seawater with DTPA; kM (M−1 s−1), metal (Cu, Fe, and Mn) reaction rate constant in seawater calculated from regression analysis of the observed first order decay rates; kobs (s−1), from the metal addition experiments. The rate of reaction with organic matter is assumed to be equal to the reaction in the presence of DTPA, korg = kDTPA. Calculation of Superoxide Decay Rate kSW in Seawater from Analysis. The measured values of the first order decay rate kobs from each experiment were combined using the following formula including all the experimental parameters:

EXPERIMENTAL SECTION Complete descriptions of the experimental methods can be found in the Supporting Information (SI) accompanying this paper. Materials and Chemicals. Due to the low concentrations of the trace metals present in seawater, all laboratory work was performed in a trace metal clean chemistry laboratory under ISO Class 5 conditions. For shipboard work a specially designed containerized clean room (Clean Modules U.K.) belonging to the GEOMAR was employed. The reagents were prepared as for our earlier works on O2− reactivity in seawater, a complete description is provided in the SI accompanying this paper. All chemicals that were used in this study were of ultrapure grade unless noted. Ultrapure (UP) water (resistivity > 18.2 MΩ cm−1) was obtained in the laboratory and in the ship going clean container via a Millipore Synergy 185 system that was fed by an Elix-3 (Millipore) reverse osmosis system connected to the mains supply. All plasticware and bottles (low density high polyethylene (LDPE) and poly(tetrafluoroethylene) (PTFE)) were extensively cleaned according to the GEOTRACES trace metal clean protocols.21 Field Sample Collection. Seawater samples were collected at 6 GoFlo-Stations occupied during the Meteor M83/1 expedition in the ETNA (Table S1 and Figure S2, SI). Procedure To Determine Superoxide Reactivity in Seawater. An aliquot of 60 mL of clean seawater was transferred directly into a PTFE bottle. This aliquot was either left unamended or diethylenetriaminepentaacetic acid (DTPA), Cu (0.79, 1.58 nM), Mn (1.00, 2.00 nM), or Fe (0.90, 1.79 nM) was added and equilibrated for at least 12 h. Subsequently 10 mL of the prepared seawater aliquots was pipetted accurately into 60 mL PTFE bottles. Twenty microliters of an 18 μM KO2 solution was added directly when the

kobs = korg + k Cu[Cu] + kFe[Fe] + kMn[Mn]

(1)

There are a number of assumptions inherent in this approach: (1) The response of the system is overall first order with respect to O2− and that the rate constant (kobs) can be described as a linear combination of the organic and metal species involved (e.g., as described above in eq 1). Furthermore to satisfy the conditions for multiple linear regression (MLR) it assumes that each variable is linearly independent and there is no collinearity. The approach also assumes that is there are no 10250

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

Figure 2. Example of the contribution of the organic matter (gray) and the three trace metals to ksw with data from GoFlo-Station4, 400 m (pie chart). The contribution of Mn (red), Cu (blue), and Fe (green) is assessed via titration which is shown separately as x−y plots for each metal.

synergistic effects (e.g., Cu(I) reduces Fe(III) resulting in kFe being dependent on the Cu concentration). (2) All the metal species are made inert by complexation with DTPA, leaving only reactions with organic species and the uncatalysed selfdismutation reaction as the pathways for O2− decay. (3) The self-dismutation reaction is significantly small that it can be ignored. In practice, as this value is reasonably well described in seawater as a function of pH,6,26 we include it in the fitting procedure or solve directly.6 (4) There are no other species involved in the dismutation of O2− in seawater. (5) The value of the metal reaction rate with O2−, kM, represents a catalytic reaction where the metal is cycled between the reduced and oxidized states via reaction with O2− such that the overall reaction rate is kM = 2kredkox/(kred + kox), where kred is the reaction rate of the reduced species with O2− and kox is the reaction rate of the oxidized species with O2−. Equation 1 was solved by combining the data from the eight independent experiments and the measured dissolved metal concentrations and solved for the four unknowns using the least-squares MLR analysis tool in Excel. Individual metal addition experiments were also solved using single factor regression analysis as previously described. Figure 1 shows the four possible cases considered for the decay of O2− in seawater for the metal addition experiments. All four cases are shown in the same way: the observed O2− loss rate in the presence of DTPA (kDTPA) is assumed to be at zero concentration, the observed rate in unamended seawater, kSW, at its natural metal concentrations (measured by graphite furnace atomic absorption (ETAAS)) and seawater with two increasing metal (M) additions (nM). Case 1a represents a scenario when there is a linear response for the metal additions and the intercept, kint, is equivalent to kDTPA indicating that the reaction in seawater can be described solely by a single metal and the organic reaction (korg). Case 1b is a special case related to case 1a in which there is no apparent organic reaction (korg = 0) and the selfdismutation reaction becomes significant. A further possibility is

Case 2, where the intercept of the individual metal additions results in a value below kDTPA; this represents a situation where some of the metal is inert to exchange with DTPA. The final possible scenario (Case 3) considered is where multiple metals contribute to the signal and the intercept for the single metal titration lies above the value for kDTPA. On the basis of our previous work in the Southern Ocean we expected that most of our data would fit Case 3.

■

RESULTS The locations of the Go-Flo sampling stations in the ETNA are shown in Figure S2 and listed in Table S1 in the SI. In this paper we will focus on three stations: station 3 on the Mauritanian shelf and stations 4 and 6 in the open ocean. Stations 3 and 6 are located in upwelling zones, on the Mauritanian shelf and the equator, respectively. For a full description of the study sites, the reader is referred to other recent works on this region.27,28 There were distinct differences in the phytoplankton communities between the upwelling zones with the picoplankton Prochlorococcus dominating the biomass (∼80% of total chlorophyll a in the upper 60 m) at station 6, while it was almost completely absent at station 3 on the Mauritanian shelf which had high concentrations of diatoms and haptophytes in the mixed layer (taxonomic chlorophyll data courtesy of Jasmin Franz, GEOMAR). Station 4 also had high concentrations of Prochlorococcus present throughout the euphotic zone (∼60% of total chlorophyll a in upper 60 m), and a distinct chlorophyll maximum was found at 50 m. The cyanobacterium Synechococcus was present in the upper part of the water column at station 3 but was found only in low concentrations at stations 4 and 6. The dissolved metal concentrations (Table S2, SI) were similar to concentrations and distributions observed in the ETNA previously.29−31 Cu increased monotonically with depth at most stations, and concentrations ranged from 0.5 to 1 nM with the highest values found at station 3 near the Mauritanian 10251

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

shelf. Fe concentrations increased with depth throughout the water column at all stations, though at stations 4 and 6 the profiles showed a distinct minimum in the vicinity of the chlorophyll maximum. Highest concentrations of Fe were found in the bottom waters at station 3, consistent with the input of resuspended sediment at that location as has been observed recently for particulate Fe along this shelf region.32 Dissolved Mn concentrations were elevated in the mixed layer of the ETNA with maxima of 2−3 nM due to atmospheric Saharan dust inputs and upwelling as has been observed previously.30 For open ocean stations Mn decreased rapidly in the euphotic zone to deep values of 0.3 nM as found in earlier studies.31 Figure 2 shows an example of data from 400 m at station 4 of the individual metal additions and the resulting estimation, by MLR, displayed as a pie chart of the contribution of organic matter and Fe, Cu, and Mn toward the decay rate of O2−. In this example it can be seen that all three metals contribute to the loss, but Mn to a greater extent than Fe and Cu, and this example is analogous to Case 3 as shown in Figure 1. In general, most data could be described by Case 3, suggesting more than one metal played a role in the decay of O2−. For some samples from surface waters with high Mn concentrations the data could not be statistically distinguished (t test, P = 0.05) from Case 1. We observed a few examples of Case 2 data that were statistically significant (i.e., kint < kDTPA) for Mn in the upper water 100 m at station 3 on the Mauritanian shelf and in the open ocean at station 4 at 60 m. The linearity of the increase in kobs in response to metal additions can be used as an indicator of a catalytic reaction occurring when the catalyst; in this case the metal added reacts with more than a stoichiometric amount of O2−,6 and this appears to be the case in the present work. This is best exemplified by the case of the Mn additions for which the response was nearly always linear in apparent agreement with the lack of organic complexation33 (i.e., Mn present as the free aquo ion). Additions of Cu in most cases resulted in a clear linear response to the additions, as we had observed previously,6,17 and this most likely reflects the presence of a large excess of Cu complexing ligand in seawater.34 Additions of Fe typically saw either no statistically significant change in kobs or only a small linear increase; this may indicate that the added Fe is bound by soluble or colloidal ligands that are relatively inert to reaction with O2−.17 The lack of a response in the case of Fe does not invalidate our assumptions list above as it simply results in a kFe below our detection limit (1 × 106 M s−1) and then the Fe term effectively drops out of eq 1. The complete compilation of elemental concentrations and calculated rate constants at each station and depth for this work are found in Tables S3 and S4 of the SI, respectively. A useful way to observe the contributions to superoxide decay from each pathway is to plot the results of the MLR for the ambient seawater conditions (i.e., no additions) as a pie graph in which the area of each slice of pie is proportional to the contribution to the decay of O2−. In Figures 3−5 we show the vertical distribution of each candidate process as a function of depth for GoFlo-Stations 3, 4, and 6 by plotting each result as a pie graph; in this way the vertical extent and influence of each process can be readily assessed. Please note that this approach precludes the display of error estimates but these values can be found in Table S6 of the SI. Overall, the calculated rates, kSW‑calc (as determined using the coefficients found in the MLR at the ambient metal concentrations),

Figure 3. Depth profile of observed loss rate of O2−, kobs, at the GoFloStation 3. The reaction with organic matter, korg, is assumed to be equal to the measured reaction when DTPA is added, kDTPA (beige line). kSW is the reaction measured in unamended seawater (blue line). kSW‑calc is the calculated reaction with seawater (red line). With the help of the pie charts, the contribution from organic matter, Cu, Fe, and Mn on ksw‑cal is shown as explained for Figure 2.

coincide generally well with the measured rate, kSW (blue). In the open ocean Mn clearly dominates the reaction in the euphotic zone (Figures 4 and 5) in contrast to the shelf

Figure 4. Contribution from organic matter, Cu, Fe, and Mn at GoFlo-Station 4 and plotted in the same way as for Figure 3.

(GoFlo-Station 3, Figure 3) where organic material is the main contributor throughout the whole water column, but most pronounced in the surface, where also high rates for kDTPA were observed. At the offshore stations the kDTPA rates varied little with depth, indicating a relatively constant contribution from organic matter to the decomposition of O2−.

■

DISCUSSION Mn Reactivity with O2− in Seawater. Mn speciation in seawater contrasts with that of Fe and Cu as there is no apparent organic complexation,33 and it exists mainly as the free aquo ion, MnCl+, and to a smaller extent as sulfate and carbonate complexes.35 Mn is similar to Fe in that the higher oxidation states are poorly soluble; however, there is no strong evidence for organic forms of Mn(III) or Mn(IV). The abiotic 10252

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

ococcus was the dominant phytoplankton; thus one possible explanation is that the Cu complex produced by Prochlorococcus reacts more slowly with O2−, though this remains to be proved. In this study the O2− decay pathway with Fe was a minor component throughout the study area in agreement with our previous work in the ETNA close to Cape Verde17 and in the Southern Ocean.6 It appears that much of the Fe present is relatively inert to reaction with O2− and may exist as soluble or colloidal organic complexes.17,48 Recent work suggests that the slow kinetics of exchange of Fe in thermodynamically weak Fe binding ligands helps to maintain soluble Fe in this region.49 It is now apparent that the O2− reactivity of a complex is not simply related to the conditional or thermodynamic binding strength of the complex. For example the Cu complex of 1,4,8,11-tetraazacyclotetradecane (Cyclam) (log KCu = 27.2) is thermodynamically strong and essentially inert to exchange in seawater,50 but SOD active, because it has free sites for water exchange.51 Measurement of the rate of exchange of fluoride (19F) with SODs has shown strong correlations with their O2− reaction rates.52,53 These findings have been suggested as evidence for an inner sphere second-order mechanism where ligand exchange is the rate limiting step and the dissociation of the complex is not required.54 Thus it is evident that specific data on the O2− reactivity of specific Cu and Fe complexes and isolated from cultures of relevant marine bacteria and phytoplankton would help to interpret field data of this type. In the companion paper to this work40 it is demonstrated that multicollinearity does not reduce the predictive power or reliability of the model as a whole, at least within the sample data themselves; it only affects calculations regarding individual predictors. O2− Loss Rates in the Presence of DTPA: Organic or Nonexchangeable SODs. At stations in regions of high primary productivity (stations 2, 3, and 6), elevated values for kDTPA were found in the euphotic zone and may represent increases in fresh reactive organic matter produced by phytoplankton exudation or zooplankton grazing. However, measurements of bulk CDOM absorbance and fluorescence showed no direct correlation between kDTPA and CDOM as seen previously in oligotrophic waters at CVOO.8 In that work, quinones were identified as the most likely functional group to explain the vertical distribution of kDTPA, and this may be derived from bacterial cell walls. In the following we explore the possibility that kDTPA includes inert metal complexes that can act as SODs. SODs catalyze the dismutation reaction of O2− radicals to H2O2 and O2. These different metalloproteins are grouped by their central redox-active metal cofactor into three major families: Cu,Zn-SOD, Mn- or Fe-SOD (specific for either/or, or some can use both = cambialistic), and Ni-SOD.55 SODs fulfill an important role as a primary defense mechanism against oxidative stress and are found in most cells in the cytosol, chloroplasts, and mitochondria depending on the type of SOD. A review of SOD usage by algae56 indicated that FeSODs are prevalent in Archaea, which include some cambialistic SODs. The MnSOD gene, sodA, is more widely distributed in the bacteria and eukaryote in particular some marine diatoms.57 CuZnSODs appear to be mostly absent in eukaryotic algae58 but are found in the periplasm of α, β, and γ proteobacteria.56 In the marine environment, recent work has shown that NiSODs are the sole SOD found in the cyanobacteria Prochlorococcus and in some open ocean Synechococcus species.59

Figure 5. Contribution from organic matter, Cu, Fe, and Mn at GoFlo-Station 6 and plotted in the same way as for Figure 3.

oxidation of Mn(II) by O2 is slow,36,37 but the rate is enhanced on particle surfaces 36,38 and via microbial processes. 39 Assuming no organic complexation of Mn(II), it would be inferred that kMn would not vary with depth as there are only minor changes in pH and chloride or carbonate concentrations. This is indeed what we observed Table S4, (SI) with no significant variation in kMn apparent with depth and a mean value of log kMn = 7.29 ± 0.26 (n = 35). For more details on the proposed mechanism for the Mn reactions with O2− in seawater, the reader is referred to the companion paper to this work.40 As can be seen in Figures 3−5, Mn is clearly the dominant pathway for O2− decay in the upper water column in this study. In this region the high deposition fluxes of Saharan dust provides elevated concentrations of particulate Mn that partially dissolves on contact with seawater.41 Photochemical activity in this region is high leading to high levels of H2O242 in the euphotic zone that reacts with particulate MnO2 helping to maintain elevated levels of Mn(II) in the dissolved phase.43,44 In our earlier work in the low Mn Southern Ocean6 we found that Cu was the dominant sink for O2−; in the present case the higher concentrations of Mn effectively outcompete Cu for O2− leading to an increase in the overall rate of O2− loss and a significant shift in the reaction pathway. O2− Reactions with Cu and Fe. In our previous work in the Southern Ocean6 and in the Tropical Atlantic near Cape Verde17 we had observed that Cu was the major pathway for O2− decomposition. In the present work we found that reactions with Mn and CDOM reduced the overall significance of the Cu pathway in the ETNA, compared to the Southern Ocean,6 but that also we observed lower values of kCu than before (Table S5, SI). A possible explanation for the latter result is the stronger apparent Cu complexation in these waters, as it has been shown that strong Cu binding ligands are produced by both Synechococcus and Prochlorococcus45 which appear to be effluxed from the cell in response to Cu stress.46 Other eukaryotic phytoplankton also produce Cu binding ligands in response to Cu stress but these ligands are typically weaker Cu complexing agents.47 The reactivity of O2− with the Cu in the Cu complexes produced by Synechococcus in culture has been determined, kCu = 5 ± 3 × 107 M−1 s−1.2 There are no reported values for kCu for Cu binding ligands produced by other phytoplankton species. Throughout our study, Prochlor10253

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

M83/1 dissolved Nickel (Ni) concentrations ranged from 2 to 6 nM (Table S3, SI) in the water column. The only study published on Ni speciation in the open ocean suggests that only 10−20% of the dissolved Ni is present as strong organic complexes, with the rest present as weak organic or inorganic species.69 Thus potentially 0.2−1.2 nM of Ni may have been present as Ni organic complexes, some of which may have been Ni-SODs or other extracellular Ni enzymes or transport ligands.70 However the slow exchange kinetics of Ni71 makes it difficult to determine the true degree of organic complexation in seawater as most approaches measuring its speciation are most likely of shorter duration than that required to reach equilibrium. In relation to the present work, as the inorganic form of Ni(II) is apparently in excess in seawater and by itself it does not react rapidly with O2−, it would be expected that metal addition experiments with inorganic Ni(II) would not influence kSW. Thus the question of whether Ni-SODs are present in solution is still open, and further work on this is needed. Cobalt (Co) when incorporated into Vitamin B12cob(II)alamin has been identified as a functional SOD;72 however, dissolved cobalt (all forms) is only present at 40−100 pM in the Atlantic Ocean.73 If all the Co were B12 it would only be a minor contributor to kSW (kCo = 107 M−1 s−1 × 10−10 M = 10−3 s−1). Most other trace metals lack a suitable redox pair for reaction with O2− (e.g., Cd and Zn) or their reactivity (e.g., Mo and V) with O2− is sluggish.74 The present work strongly indicates that reaction with dissolved Mn(II) can be a significant pathway for O2− decay in the surface ocean of regions impacted by terrestrial runoff, sediment resuspension, or atmospheric deposition.

Extracellular SODs are known to be produced by higher organisms in the terrestrial environment,60 but there is no data on marine organisms. There is, however, evidence for the presence of dissolved catalase or peroxidase in seawater, as experiments with filtered seawater (0.4 μm) indicate a continual loss of H2O2 in the dark.15 Ultrafiltration experiments suggest that the decomposition of H2O2 is predominantly by colloidal species (1 kDa < x < 0.4 μm);61 thus specific SODs from bacteria and phytoplankton may be present in the dissolved phase through direct excretion, zooplankton grazing, or viral lysis. The question then arises, can natural SODs contribute to the high values for kDTPA found in the euphotic zone (Figures 3−5), and can they exchange their metal content with DTPA? Unfortunately there is very little data about the binding strength of the metal centers in SODs. It is common practice in biological studies to add EDTA at a concentration of 0.1 mM immediately prior to experiments using Cu/Zn SOD to bind extraneous Cu. It is reported that EDTA has no influence on the observed O2− decay rate,62 though these experiments are too short to allow the Cu in the SOD to come into equilibrium with EDTA. Some data does exist for the Cu/Zn SOD binding strength for bovine erythrocyte SOD Cu2Zn2SOD2, which contains two identical subunits. For Cu there was a strong pH dependence with log K1 = 15.6 and log K2 = 14.1 at pH 7 (0.2 M phosphate buffer), and the solutions took several days to come into equilibrium with 2-pyridinecarboxylate.63 Interestingly it has also been suggested that DTPA can bind to Cu (in pH 7.2 50 mM sodium phosphate buffer) at the active site of bovine erythrocyte CuZnSOD without removing the Cu from the SOD nor with any resultant change in the rate of O2− decay.64 While it is speculation to suggest that this result can be extended to other SODs it does suggest that DTPA may not remove the entire metal signal. We could find no relevant data on the binding strength of Mn- or Fe-SODs. There have been a few studies on the use of EDTA to remove metals from organic ligands in seawater. EDTA was used in combination with XAD-2 resin65 or electrodialysis66 to determine the binding strength of natural Cu and Zn complexes in seawater. Kinetic experiments indicated that equilibrium was achieved in less than 24 h under the experimental conditions employed, though it is not clear if 100% recovery was achieved.65,66 In our work, Case 2 above would result from incomplete exchange with DTPA, and we only observed this in the specific case of Mn. This could suggest the existence of small concentrations of Mn-SODs in the dissolved phase. As Mn-SODs react significantly more rapidly with O2− (log kMnSOD ∼ 9.3)67,68 than Mn2+ (log kMn ∼7.2), the Case 2 situation shown in Figure 1 then the decay rate of 1 nM of Mn2+ (kobs = 0.016 s−1) would be equivalent to only 10 pM of a putative MnSOD. Such a low concentration of a Mn organic complex would be extremely difficult to determine analytically in the presence of nanomolar inorganic Mn. Thus while it appears that DTPA removes the bulk of metals from the natural organic complexes, further work is required to verify this for SODs from the marine environment in light of the Ye et al. work.64 Involvement of Other Metals as SODs in Solution. NiSODs are present in Synechococcus and Prochlorococcus,59 the predominant picoplankton found in the ETNA.27 Thus there is the possibility that Ni-SODs may be released to the dissolved phase via grazing or viral lysis; the stability of extracellular NiSODs in seawater is not known at present, however. During

■

ASSOCIATED CONTENT

S Supporting Information *

Full details of station locations and the experimental setup and related analysis methods are supplied as Supporting Information. The cruise track and stations locations (Figure S2), experimental set up and design (Figure S4) are also provided. Examples of the signal output from the FeLume have previously been published.6,8 This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The officers and crew of the research vessel Meteor are gratefully acknowledged for their help in performing sampling at sea. This work was supported by the efforts of Peter Streu, Mirja Dunke,r and Dörte Nitschkowski (all GEOMAR) in the laboratory in Kiel. Special thanks are due to Jasmin Franz (GEOMAR) for providing the HPLC data. This work is a contribution of the Collaborative Research Centre 754 “Climate - Biogeochemistry Interactions in the Tropical Ocean” (www.sfb754.de), which is supported by the Deutsche Forschungsgemeinschaft (DFG). K.W.’s participation was financially supported by a grant, awarded to P.L.C., from the DFG (CR145/17-1). Financial support for M.I.H. was provided by the BMBF Verbundprojekt SOPRAN (FKZ03F0462A and 10254

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

(20) Goldstone, J. V.; Voelker, B. M. Chemistry of superoxide radical in seawater: CDOM associated sink of superoxide in coastal waters. Environ. Sci. Technol. 2000, 34, 1043−1048. (21) Cutter, G. A.; Andersson, P.; Codispoti, L.; Croot, P. L.; Francois, R.; Lohan, M.; Obata, H.; Rutgers van der Loeff, M. Sampling and sample-handling protocols for GEOTRACES cruises. http://www.geotraces.org/libraries/documents/Intercalibration/ Cookbook.pdf (22) Levenberg, K. A method for the solution of certain non-linear problems in least squares. Quart. Appl. Math. 1944, 2, 164−168. (23) Marquardt, D. W. An algorithm for least-squares estimation of nonlinear parameters. SIAM J. Appl. Math. 1963, 11 (2), 431−441. (24) Press, W. H.; Teukolsky, S. A.; Vettering, W. T.; Flannery, B. P., Numerical Recipes in C: the art of scientific computing, 2nd ed.; Cambridge University Press: Cambridge, 1992. (25) Miller, J. C.; Miller, J. N., Statistics for Analytical Chemistry, 3rd ed.; Ellis Horwood: London, 1993. (26) Zafiriou, O. C. Chemistry of superoxide ion (O2−) in seawater. I. pKasw* (HOO) and uncatalysed dismutation kinetics studied by pulse radiolysis. Mar. Chem. 1990, 30, 31−43. (27) Tarran, G. A.; Heywood, J. L.; Zubkov, M. V. Latitudinal changes in the standing stocks of nano- and picoeukaryotic phytoplankton in the Atlantic Ocean. Deep Sea Res., Part II 2006, 53 (14−16), 1516−1529. (28) Heller, M. I.; Gaiero, D.; Croot, P. L. Basin scale survey of marine humic fluorescence in the Atlantic: relationship to iron solubility. Global Biogeochem. Cycles 2012, 27 (1), 88−100. (29) Croot, P. L.; Streu, P.; Baker, A. R. Short residence time for iron in surface seawater impacted by atmospheric dry deposition from Saharan dust events. Geophys. Res. Lett. 2004, 31, L23S08 DOI: 10.1029/2004GL020153. (30) Pohl, C.; Croot, P. L.; Hennings, U.; Daberkow, T.; Budeus, G.; Loeff, M. R. v. d. Synoptic transects on the distribution of trace elements (Hg, Pb, Cd, Cu, Ni, Zn, Co, Mn, Fe, and Al) in surface waters of the Northern- and Southern East Atlantic. J. Mar. Sys. 2011, 84 (1−2), 28−41. (31) Statham, P. J.; Yeats, P. A.; Landing, W. M. Manganese in the eastern Atlantic Ocean: processes influencing deep and surface water distributions. Mar. Chem. 1998, 61 (1−2), 55−68. (32) Lam, P. J.; Ohnemus, D. C.; Marcus, M. A. The speciation of marine particulate iron adjacent to active and passive continental margins. Geochim. Cosmochim. Acta 2012, 80 (0), 108−124. (33) Sunda, W. G. Measurement of manganese, zinc and cadmium complexation in seawater using Chelex ion exchange equilibria. Mar. Chem. 1984, 14 (4), 365−378. (34) Moffett, J. W. Temporal and spatial variability of copper complexation by strong chelators in the Sargasso Sea. Deep Sea Res. 1995, 42, 1273−1295. (35) Byrne, R. H. Seawater trace metal speciation. App. Geochem. 1988, 3 (1), 85. (36) Morgan, J. J. Kinetics of reaction between O2 and Mn(II) species in aqueous solutions. Geochim. Cosmochim. Acta 2005, 69 (1), 35. (37) von Langen, P. J.; Johnson, K. S.; Coale, K. H.; Elrod, V. A. Oxidation kinetics of manganese (II) in seawater at nanomolar concentrations. Geochim. Cosmochim. Acta 1997, 61 (23), 4945−4954. (38) Yeats, P. A.; Strain, P. M. The oxidation of manganese in seawater: Rate constants based on field data. Estuar. Coast. Shelf. Sci. 1990, 31 (1), 11−24. (39) Tebo, B. M.; Johnson, H. A.; McCarthy, J. K.; Templeton, A. S. Geomicrobiology of manganese(II) oxidation. Trends in Microbiol. 2005, 13 (9), 421−428. (40) Wuttig, K.; Heller, M. I.; Croot, P. L. Reactivity of Mn(II/III)desferrioxamine B with O2, O2−, and H2O2 in seawater. Environ. Sci. Technol. 2013, DOI: 10.1021/es4016606. (41) Mendez, J.; Guieu, C.; Adkins, J. Atmospheric input of manganese and iron to the ocean: Seawater dissolution experiments with Saharan and North American dusts. Mar. Chem. 2010, 120 (1−4), 34−43.

03F0611A), and this forms part of the German contribution to SOLAS (Surface Ocean Lower Atmosphere Studies).

■

REFERENCES

(1) Voelker, B. M.; Sedlak, D. L. Iron reduction by photoproduced superoxide in seawater. Mar. Chem. 1995, 50, 93−102. (2) Voelker, B. M.; Sedlak, D. L.; Zafiriou, O. C. Chemistry of superoxide radical in seawater: Reactions with organic Cu complexes. Environ. Sci. Technol. 2000, 34, 1036−1042. (3) Zafiriou, O. C.; Voelker, B. M.; Sedlak, D. L. Chemistry of the superoxide radical (O2−) in seawater: Reactions with inorganic copper complexes. J. Phys. Chem. A 1998, 102 (28), 5693−5700. (4) Rose, A. L.; Waite, D. Role of superoxide in the photochemical reduction of iron in seawater. Geochim. Cosmochim. Acta 2006, 70 (15), 3869−3882. (5) Hansard, S. P.; Easter, H. D.; Voelker, B. M. Rapid reaction of nanomolar Mn(II) with Superoxide radical in seawater and simulated freshwater. Environ. Sci. Technol. 2011, 45 (7), 2811−2817. (6) Heller, M. I.; Croot, P. L. Superoxide decay kinetics in the Southern Ocean. Environ. Sci. Technol. 2010, 44 (1), 191−196. (7) Heller, M. I.; Croot, P. L. Application of a superoxide (O2−) thermal source (SOTS-1) for the determination and calibration of O2− fluxes in seawater. Anal. Chim. Acta 2010, 667, 1−13. (8) Heller, M. I.; Croot, P. L. Kinetics of superoxide reactions with dissolved organic matter in tropical Atlantic surface waters near Cape Verde (TENATSO). J. Geophys. Res. 2010, 115 (C12), C12038. (9) Micinski, E.; Ball, L. A.; Zafiriou, O. C. Photochemical oxygen activation - superoxide radical detection and production-rates in the eastern Caribbean. J. Geophys. Res.-Oceans 1993, 98 (C2), 2299−2306. (10) Marshall, J.-A.; Hovenden, M.; Oda, T.; Hallegraeff, G. M. Photosynthesis does influence superoxide production in the ichthyotoxic alga Chattonella marina (Raphidophyceae). J. Plankton Res. 2002, 24 (11), 1231−1236. (11) Kim, D.; Oda, T.; Ishimatsu, A.; Muramatsu, T. Galacturonic acid-induced increase of superoxide production in red tide phytoplankton Chattonella marina and Heterosigma akashiwo. Biosci. Biotechnol. Biochem. 2000, 64, 911−914. (12) Rose, A. L.; Godrant, A.; Furnas, M.; Waite, T. D. Dynamics of nonphotochemical superoxide production and decay in the Great Barrier Reef lagoon. Limnol. Oceanogr. 2010, 55 (4), 1521−1536. (13) Rose, A. L.; Webb, E. A.; Waite, T. D.; Moffett, J. W. Measurement and implications of nonphotochemically generated superoxide in the equatorial Pacific Ocean. Environ. Sci. Technol. 2008, 42 (7), 2387−2393. (14) Hansard, S. P.; Vermilyea, A. W.; Voelker, B. M. Measurements of superoxide radical concentration and decay kinetics in the Gulf of Alaska. Deep Sea Res., Part I 2010, 57 (9), 1111−1119. (15) Vermilyea, A. W.; Hansard, S. P.; Voelker, B. M. Dark production of hydrogen peroxide in the Gulf of Alaska. Limnol. Oceanogr. 2010, 55, 580−588. (16) Nakano, M.; Sugioka, K.; Ushijima, Y.; Goto, T. Chemiluminescence probe with Cypridina luciferin analog, 2-methyl-6-phenyl3,7-dihydroimidazo[1,2-a]pyrazin-3-one, for estimating the ability of human granulocytes to generate O2. Anal. Biochem. 1986, 159 (2), 363−369. (17) Heller, M. I.; Croot, P. L. Superoxide decay as a probe for speciation changes during dust dissolution in tropical Atlantic surface waters near Cape Verde. Mar. Chem. 2011, 126 (1−4), 37−55. (18) Croot, P. L.; Laan, P.; Nishioka, J.; Strass, V.; Cisewski, B.; Boye, M.; Timmermans, K.; Bellerby, R.; Goldson, L.; de Baar, H. J. W. Spatial and temporal distribution of Fe(II) and H2O2 during EISENEX, an open ocean mesoscale iron enrichment. Mar. Chem. 2005, 95, 65−88. (19) Fujii, M.; Ito, H.; Rose, A. L.; Waite, T. D.; Omura, T. Superoxide-mediated Fe(II) formation from organically complexed Fe(III) in coastal waters. Geochim. Cosmochim. Acta 2008, 72 (24), 6079−6089. 10255

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256

Environmental Science & Technology

Article

(64) Ye, M.; English, A. M. Binding of polyaminocarboxylate chelators to the active-site copper inhibits the GSNO-reductase activity but not the superoxide dismutase activity of Cu,Zn-superoxide dismutase. Biochemistry 2006, 45 (42), 12723−12732. (65) Hirose, K.; Dokiya, Y.; Sugimura, Y. Determination of conditional stability constants of organic copper and zinc complexes dissolved in seawater using ligand exchange method with EDTA. Mar. Chem. 1982, 11 (4), 343−354. (66) Midorikawa, T.; Tanoue, E.; Sugimura, Y. Determination of complexing ability of natural ligands in seawater for various metal ions using ion selective electrodes. Anal. Chem. 1990, 62 (17), 1737−1746. (67) Bull, C.; Niederhoffer, E. C.; Yoshida, T.; Fee, J. A. Kinetic studies of superoxide dismutases: properties of the manganesecontaining protein from Thermus thermophilus. J. Am. Chem. Soc. 1991, 113 (11), 4069−4076. (68) Abreu, I. A.; Cabelli, D. E. Superoxide dismutases: a review of the metal-associated mechanistic variations. Biochim. Biophys. Acta 2010, 1804 (2), 263−274. (69) Achterberg, E. P.; Van Den Berg, C. M. G. Chemical speciation of chromium and nickel in the western Mediterranean. Deep Sea Res. Part II 1997, 44 (3−4), 693. (70) Dupont, C. L.; Buck, K. N.; Palenik, B.; Barbeau, K. Nickel utilization in phytoplankton assemblages from contrasting oceanic regimes. Deep Sea Res. Part II 2010, 57 (4), 553−566. (71) Hudson, R. J. M.; Morel, F. M. M. Trace metal transport by marine microorganisms: implications of metal coordination kinetics. Deep Sea Res. 1993, 40, 129−150. (72) Suarez-Moreira, E.; Yun, J.; Birch, C. S.; Williams, J. H. H.; McCaddon, A.; Brasch, N. E. Vitamin B12 and redox homeostasis: Cob(II)alamin reacts with superoxide at rates approaching superoxide dismutase (SOD). J. Am. Chem. Soc. 2009, 131 (42), 15078−15079. (73) Saito, M. A.; Moffett, J. W. Temporal and spatial variability of cobalt in the Atlantic Ocean. Geochim. Cosmochim. Acta 2002, 66 (11), 1943−1953. (74) Bielski, B. H. J.; Cabelli, D. E.; Arudi, R. L.; Ross, A. B. Reactivity Of HO2/O2− radicals in aqueous-solution. J. Phys. Chem. Ref. Data 1985, 14 (4), 1041−1100.

(42) Croot, P. L.; Streu, P.; Peeken, I.; Lochte, K.; Baker, A. R. Influence of the ITCZ on H2O2 in near surface waters in the equatorial Atlantic Ocean. Geophys. Res. Lett. 2004, 31, L23S04 DOI: 10.1029/ 2004GL020154. (43) Sunda, W. G.; Huntsman, S. A.; Harvey, G. R. Photoreduction of manganese oxides in seawater and its geochemical and biological implications. Nature 1983, 301 (5897), 234−236. (44) Sunda, W. G.; Huntsman, S. A. Photoreduction of manganese oxides in seawater. Mar. Chem. 1994, 46 (1−2), 133−152. (45) Moffett, J. W.; Brand, L. E. Production of strong, extracellular Cu chelators by marine cyanobacteria in response to Cu stress. Limnol. Oceanogr. 1996, 41, 388−395. (46) Croot, P. L.; Karlson, B.; van Elteren, J. T.; Kroon, J. J. Uptake and efflux of 64Cu by the marine cyanobacterium Synechococcus (WH7803). Limnol. Oceanogr. 2003, 48, 179−188. (47) Croot, P. L.; Moffett, J. W.; Brand, L. Production of extracellular Cu complexing ligands by eucaryotic phytoplankton in response to Cu stress. Limnol. Oceanogr. 2000, 45, 619−627. (48) Fujii, M.; Rose, A. L.; Waite, T. D.; Omura, T. Superoxidemediated dissolution of amorphous ferric oxyhydroxide in seawater. Environ. Sci. Technol. 2006, 40 (3), 880−887. (49) Croot, P. L.; Heller, M. I., The importance of kinetics and redox in the biogeochemical cycling of iron in the surface ocean. Front. Microbiol. 2012, 3, doi 10.3389/fmicb.2012.00219. (50) Croot, P. L.; Moffett, J. W.; Luther, G. W. Polarographic determination of half-wave potentials for copper-organic complexes in seawater. Mar. Chem. 1999, 67 (3−4), 219−232. (51) Kimura, E.; Sakonaka, A.; Nakamoto, M. Superoxide dismutase activity of macrocyclic polyamine complexes. Biochim. Biophys. Acta 1981, 678 (2), 172−179. (52) Viglino, P.; Rigo, A.; Stevanato, R.; Ranieri, G. A.; Rotilio, G.; Calabrese, L. The binding of fluoride ion to bovine cuprozinc superoxide dismutase as studied by 19F magnetic relaxation. J. Magn. Reson. 1979, 34 (2), 265−274. (53) Summers, J. S.; Baker, J. B.; Meyerstein, D.; Mizrahi, A.; Zilbermann, I.; Cohen, H.; Wilson, C. M.; Jones, J. R. Measured rates of fluoride/metal association correlate with rates of superoxide/metal reactions for FeIIIEDTA(H2O)− and related complexes. J. Am. Chem. Soc. 2008, 130 (5), 1727−1734. (54) Graf, E.; Mahoney, J. R.; Bryant, R. G.; Eaton, J. W. Ironcatalyzed hydroxyl radical formation. Stringent requirement for free iron coordination site. J. Biol. Chem. 1984, 259 (6), 3620−4. (55) Miller, A.-F. Superoxide dismutases: active sites that save, but a protein that kills. Curr. Opin. Chem. Bio. 2004, 8 (2), 162−168. (56) Wolfe-Simon, F.; Grzebyk, D.; Schofield, O.; Falkowski, P. G. The role and evolution of superoxide dismutases in algae. J. Phycol. 2005, 41 (3), 453−465. (57) Wolfe-Simon, F.; Starovoytov, V.; Reinfelder, J. R.; Schofield, O.; Falkowski, P. G. Localization and role of manganese superoxide dismutase in a marine diatom. Plant Physiol. 2006, 142 (4), 1701− 1709. (58) Asada, K.; Kanematsu, S.; Uchida, K. Superoxide dismutases in photosynthetic organisms: Absence of the cuprozinc enzyme in eukaryotic algae. Arch. Biochem. Biophys. 1977, 179 (1), 243. (59) Dupont, C. L.; Neupane, K.; Shearer, J.; Palenik, B. Diversity, function and evolution of genes coding for putative Ni-containing superoxide dismutases. Environ. Microbio. 2008, 10 (7), 1831−1843. (60) Fridovich, I. Superoxide radical and superoxide dismutases. Annu. Rev. Biochem. 1995, 64, 97−112. (61) Yuan, J.; Shiller, A. M. The distribution of hydrogen peroxide in the southern and central Atlantic ocean. Deep Sea Res. Part II 2001, 48, 2947−2970. (62) Goldstein, S.; Fridovich, I.; Czapski, G. Kinetic properties of Cu,Zn-superoxide dismutase as a function of metal contentOrder restored. Free Radical Biol. Med. 2006, 41 (6), 937−941. (63) Hirose, J.; Ohhira, T.; Hirata, H.; Kidani, Y. The pH dependence of apparent binding constants between apo-superoxide dismutase and cupric ions. Arch. Biochem. Biophys. 1982, 218 (1), 179−186. 10256

dx.doi.org/10.1021/es401658t | Environ. Sci. Technol. 2013, 47, 10249−10256